USA lawyers got fined $5000 for citing fictitious legal authorities produced by ChatGPT
A U.S. judge has levied a $5,000 fine against two lawyers and their law firm, Levidow, Levidow & Oberman, after they submitted falsified citations in a court filing, which were generated by the ChatGPT artificial intelligence tool.
The attorneys, Steven Schwartz and Peter LoDuca, admitted to referencing six fabricated cases in a legal brief against Avianca, a Colombian airline. While the judge acknowledged the legitimacy of using AI for legal assistance, he stressed the lawyers’ responsibility to ensure the accuracy of their submissions.
Levidow, Levidow & Oberman contested the ruling, attributing their error to misplaced trust in technology. The incident underscores the importance of verifying AI-generated content before legal submissions.
Chatbots like ChatGPT, developed by OpenAI, can sometimes produce inaccurate information, leading to what is termed as “hallucinations.” These chatbots are trained on vast internet data, operating akin to predictive text tools. While they can provide human-like responses, factual errors are possible due to the predictive nature of their functioning.
Instances of inaccuracies have raised concerns, such as when Google’s Bard gave an incorrect answer about the James Webb space telescope, prompting discussions about the haste in launching AI technologies like these. The lack of transparency regarding data sources further complicates the issue.
On the legal side, it is still unclear if the mistake of the 2 lawyers are intentional to gain upper vantages in court proceeding or accidental due to laziness, reliant too much on the availability of AI.
If the initial is deemed to be correct on further investigations, the lawyers might face more severe consequences such as the removal of attorney’s status of practicing law in the USA. The judge seemed to be agreeing with this way of reasoning, stating that the lawyers and their firm “abandoned their responsibilities when they submitted nonexistent judicial opinions with fake quotes and citations created by the artificial intelligence tool ChatGPT, then continued to stand by the fake opinions after judicial orders called their existence into question.”
In return, the law firm Levidow, Levidow & Oberman responded that they made a mistake in good faith, in contrast to the intentional bad faith manner that the Court were suggesting. The representative said that due to being widely used in their law firms with the ability to gather clear, precise information from the Internet, they become too reliant on the technology and made a simple but serious mistake.
In their response, they claimed that they didn’t think that Chat GPT is capable of making false cases based on an honest request from the lawyers handling the case.
The judge noted that one of the fabricated decisions produced by the chatbot exhibited “superficial consistency with actual judicial decisions,” although other parts contained “gibberish” and were “nonsensical.” In a different written opinion, the judge dismissed the original aviation claim, citing the expiration of the statute of limitations.
***Other Articles***
– You could check Required documents of filing trademark in USA here.
– You could see What needs to be done after registering a trademark in US? here.
– You could visit here to see What is use in commerce of filing trademark in USA.
– You could also check here to see Procedure of Trademark in USA.
Contact AAA IPRIGHT: Email: [email protected]
Or sending your inquiry by filling the form:
